Proceedings of the 11th International Coral Reef Symposium, Ft. Lauderdale, Florida, 7-11 July 2008 Session number 1

Cenozoic Evolution of Larger Benthic Foraminifers: Paleoceanographic Evidence for Changing Habitats

P. Hallock1, L. Pomar2

1) University of South Florida, College of Marine Science, 140 7th Avenue S., St. Petersburg, FL 33701, USA 2) Universitat de les Illes Balears, Departament de Ciencies de la Terra, E-07122 Palma de Mallorca, Spain

Abstract. The ever-increasing treasure-trove of paleoceanographic data relating to evolving Cenozoic ocean structure, including geographic and bathymetric gradients, provides novel insights into long-term changes in environmental conditions influencing shelf, ramp and oceanic-platform habitats occupied by carbonate- producing ecosystems. Similarly, recent studies documenting the influence of internal waves on mid- and deep- shelf habitats provides equally exciting insights into previously unrecognized environmental variability experienced by organisms living in those habitats. Paleocene-Eocene photic-dependent carbonates were dominated by calcitic coralline red algae and larger benthic foraminifers (LBF), with aragonitic corals and calcareous green algae more restricted temporally and spatially. Morphologies of LBF are strongly influenced by light availability and water motion, with larger, flatter and more fragile shapes characteristic of lower light, low wave-energy environments. Since substantial LBF habitat is at middle to outer shelf or ramp depths (i.e., ~30 to ~130 m), understanding the influence of internal waves on these habitats as oceanic thermal gradients developed through the Cenozoic can provide crucial insights into evolving environmental conditions where the most diverse and highly specialized LBF biotas occurred.

Key words: Carbonate ramp, reef, internal waves, thermocline gradients, symbiosis

Introduction influence of internal waves at those depths (e.g., Understanding biological and geochemical processes Wolanski et al. 2004; Leichter et al. 2005) with the associated with modern carbonate systems is essential records of oceanic thermal gradients through the to interpreting fossil reefs and carbonate Cenozoic (e.g., Dutton et al. 2005) can provide crucial sedimentation. Recognizing the limitations of insights into how environmental conditions changed uniformitarianism is equally crucial (Pomar & for middle and outer shelf benthic habitats. Hallock 2008). Cenozoic carbonate-producing ecosystems emerged from the remnants of Cretaceous Larger Benthic Foraminifers (LBF) biotas, evolving in the warm alkaline oceans of a The LBF are an informal group of benthic Greenhouse world, then modifying as Icehouse foraminifers characterized by relative large size (e.g., conditions developed. The latter included stronger generally > 1mm and up to 6 cm in diameter – Lee latitudinal and bathymetric temperature gradients, 1998) and complex internal morphologies. Modern declining atmospheric CO2 concentrations and representatives host a rich variety of algal declining calcium concentrations and alkalinity in the endosymbionts in a relationship similar to that oceans. between corals and their zooxanthellae. By analogy, Paleocene-Eocene photic-dependent carbonates fossil LBF are interpreted to also have hosted algal were dominated by calcitic coralline red algae and symbionts (e.g., Lee 1998; Hallock 1999; and larger benthic foraminifers (LBF), with aragonitic references therein). corals and calcareous green algae more restricted Extant LBF worldwide belong to members of six temporally and spatially. In this paper, we compare long-ranging families, the Amphisteginidae, Hottinger’s (1998) synthesis of records of Alveolinidae, Nummulitidae, Peneroplidae, Rotaliidae, diversifications and extinctions of Cenozoic LBF and Soritidae (e.g., Hallock 1999). One other family, lineages in the context of published evidence for the Calcarinidae, is characteristic of the late Neogene changes in ocean circulation and thermocline in the Indo-west Pacific, where some species are structure as Icehouse conditions developed (e.g., Lear prolific producers of beach sands (e.g., Hohenegger et al. 2000). Moreover, given that much of the 2006). Amphisteginidae, Calcarinidae, Nummulitidae diversification was at middle to outer shelf depths (i.e., and Rotaliidae are all families in the order Rotaliida, ~30 to ~130 m), synthesizing emerging records of the and are characterized by relatively transparent (i.e.,

16 hyaline) calcite shells. Members of these families transparency only occurs where surface waters are typically host diatom endosymbionts. In contrast, the sediment free and have extremely low nutrient and other families are Miliolida, whose shells of randomly plankton concentrations. arranged calcite crystals covered by a veneer of brick- LBF assemblages typically characterize Pomar’s like crystals impart a porcelaneous appearance that is (2001) light-defined zones (Fig. 1). Most of the relatively opaque to light (e.g., Debenay et al. 2000). miliolid LBF, as well as most calcarinids, live The shells of many of the larger miliolids have primarily in euphotic habitats. On the Florida reef thinned outer walls to permit light into the tract, Soritidae (especially the Archaiasinae) show chamberlets. Larger miliolids host a diversity of algal depth zonation, but none are common below about 30 symbionts (e.g., Lee & Anderson 1991). The m. Worldwide, robust Amphistegina also are found Alveolinidae host diatoms. One lineage of abundantly at less than 30 m, with progressively Peneroplidae hosts rhodophyte symbionts while a thinner or smaller chambered morphologies second lineage has chlorophyte symbionts. Similarly, dominating below 30 m. Nummulitids can be found one lineage in the Soritidae hosts dinoflagellate at less than 30 m, but most dominate at mesophotic to symbionts while the second hosts chlorophytes. oligophotic depths (>30m, see Hohenegger 2005). Distributions of modern taxa of LBF are strongly Several extant nummulite species, whose shapes are influenced by light (Fig.1). Both between species and very thin and flat, have depth distributions that peak within species trends in morphology, especially shape at 70 m or deeper. and surface-to-volume ratios, reflect light intensity Water motion is another important environmental and water motion (e.g., Hohenegger 2005), both of parameter influencing LBF distributions. Both wave which influence rates of calcification (e.g., Hallock et energy and light decrease exponentially with depth, al. 1986). Thus, both conceptual (Hallock & Glenn and both influence shell morphologies similarly, 1986; Beavington-Penney & Racey 2004) and which is one reason LBF are excellent paleodepth numerical (e.g., Hallock 1987; Mateu-Vicens et al. indicators. Lenticular shapes, thicker shell walls and, 2008) models have been developed to interpret in the case of the calcarinids, stellate morphologies paleoenvironments of LBF-rich limestones. allow such foraminifers to thrive and hypercalcify in Hallock (1988, 1999) also noted that some taxa, high light, high wave-energy environments such as notably the porcelaneous Soritidae and the stellate reef flats of the Indo-Pacific. Calcarinidae, have tended to diversify “laterally”, specializing to different relatively shallow-water, higher light habitats, while modern Amphisteginidae, and Nummulitidae have diversified vertically, with robust shapes characteristic of shallower dwelling species and flat shapes characteristic of deeper depths. Historically, the terms photic or euphotic have been used to describe environments where there was sufficient light for an excess of photosynthesis over respiration (e.g., Hallock & Schlager 1986). Pomar (2001) proposed terminology to distinguish among Figure 1: Major families of larger benthic foraminifers illustrating photic habitats. He used “euphotic” only for depths shapes and general depth ranges (Hallock 1999) in the context of euphotic, mesophotic and oligophotic habitats as defined by Pomar where light is sufficient for high rates of (2001). photosynthesis and associated hypercalcification, typically less than 30 m even in very clear water. He In contrast, discoid morphologies are best adapted used “mesophotic” to characterize depths where light to quieter environments, where they either burrow is sufficient for photosynthesis to support substantial their apertural face into the surficial sediments at a calcification rates, but not true hypercalcification, i.e., low angle, leaving most of the dorsal surface exposed depths in the 20-70 m range, again dependent upon to light, or sit flat on the substratum. Either way, they water transparency. He used the term “oligophotic” to are superbly adapted to harvesting food and dissolved characterize depths where there is limited light nutrients from the substratum and pore waters, while penetration sufficient to support calcifiers like their algal symbionts capture radiant energy from coralline red algae, and very thin flat LBF and corals. sunlight. If their habitat is infrequently disturbed, e.g., As Hallock (1987) demonstrated using a modeling by storm waves or deposit-feeding animals, these approach, well developed biotas adapted to low light foraminifers can literally spend years accumulating for photosynthesis depend upon consistently high the sparse resources required for reproduction water transparency that permits light penetration to (Hallock 1985). Exceptions to this generalization for depths >70 m. Such consistently high water discoid morphologies are shallow-dwelling Soritidae

17 that commonly attach to hard or phytal substrata in optimize light capture rather than food capture (e.g., reef-flat environments. Jarrett et al. 2005). As a consequence, light energy must be relatively predictable for the investment to be Costs and benefits of algal symbiosis beneficial, requiring that water column transparency There are energetic costs to establishing and be consistently very high. maintaining symbioses between heterotrophic hosts and photoautotrophic symbionts (e.g., Hallock 1981; Stoecker 1998). Unless environmental conditions tip the energy balance in favor of benefits of symbiosis, photoautotrophic and heterotrophic taxa will predominate. The energy from photosynthesis must provide more than half the energy needed by the holobiont to be beneficial (Stoecker 1998). Hallock (1981) predicted that, under nutrient-poor conditions with sufficient light, algal symbiosis can provide literally orders of magnitude energetic advantage over autotrophic-heterotrophic strategies. She further postulated that the most advantageous conditions for algal symbiosis occur when essential nutrients are extremely scarce and the only concentrated forms available are in particulate organic carbon such as plankton, bacteria or organic detritus. Hallock (1985) examined life history strategies that would be most advantageous for LBF. She concluded that hydrodynamic environments, which tend to also Figure 2: Data from Hallock et al. (1991) showing surface-to- be high light environments, would favor faster growth, thermocline temperature and fluorescence gradients for the modern higher fecundity and relatively short life spans. In northern Caribbean, annotated to indicate changes in light and contrast, Hallock (1985, 1987) predicted that, in temperature in the upper 150 m. mesophotic to oligophotic environments where radiant energy is limited but the probability of Bathymetric gradients and internal waves disturbance by hydrodynamic events is much reduced, In modern subtropical and tropical oceans, outside of longer life spans and production of fewer, larger upwelling zones, surface waters are warm and embryons are advantageous. Mesophotic conditions generally relatively well mixed by winds down to appear to be optimal if water transparency is roughly 100 m. The base of the mixed layer/top of relatively dependable, because the holobionts are still the thermocline also corresponds to the top of the able to get substantial energy from photosynthesis, pycnocline and nutricline, as well as the chlorophyll while the potential for either photo-oxidative stress or maximum zone (Fig. 2). In addition, O2 concentration physical damage by hydrodynamic events are low. declines and pCO2 increases relatively rapidly with The greatest risk is the potential for insufficient depth. Above this layer, there is sufficient light for sunlight during turbidity events, although there is the photosynthesis but generally limited nutrients. Below potential for energy supplementation from feeding. there is minimal light, but more organic carbon and Interestingly, the two extant families most prevalent biological activity utilizing oxygen, releasing in mesophotic and oligophotic depths are the nutrients and CO2. This zone is very biologically and amphisteginids and nummulitids. The former can chemically dynamic. become dormant within 24 hours of darkness, while at The base of the mixed layer and uppermost least some of the latter are bacteriovores (Lee 1998). thermocline corresponds with deeper mesophotic to Oligophotic environments tend to be low energy, oligophotic conditions described for outer shelf and both radiant and hydrodynamic. So the limiting factor ramp biotas. Thus, changes in physical and chemical must be the tradeoff between the energy required to conditions along that gradient would strongly support the symbiosis versus the energy gain from influence deeper dwelling LBF biotas. The final photosynthesis. The morphological investment in consideration here is the influence of internal waves maintaining a symbiosis appears to be extreme in very on LBF in mesophotic and oligophotic environments. low light environments – long lives spent producing Leichter and colleagues studied internal waves in extremely large, flat shells. The strategies appear Jamaica and the Florida Keys, recording substantial similar for zooxanthellate corals in oligophotic temperature fluctuations on tidal frequencies at depths environments: significant investment in structures that below about 30 m (e.g., Leichter et al. 2005; Leichter

18 & Genovese 2006). Wolanski et al. (2004) recorded which arose in the Paleocene-Eocene, have since been temperature variations with depth in Palau, where represented through the Cenozoic, as have the strong internal waves impart as much as five degrees Soritidae, Peneroplidae and new lineages of the of temperature variability on habitats between 30 and Nummulitidae. With the Neogene closure of the 100 m depth on tidal cycles (Fig.3). circumtropical seaway and the isolation of Atlantic biotas, the archaiasine lineage of the Soritidae diversified in the Western Atlantic, while the Calcarinidae diversified in the Indo-west Pacific.

Figure 3: Data from Wolanski et al. (2004) recording internal wave activity on tidal cycles on the shelf of Palau, Western Figure 4: Comparison of the development of deep-sea temperatures Caroline Islands, annotated to show that benthic communities (Leer et al. 2000) with diversification and extinction history of living at 90 m depth experience substantial temperature fluctuations. larger benthic foraminiferal (LBF) lineages (Hottinger 1998).

As a consequence, subtropical/tropical mesophotic Given the prevalence of LBF as carbonate-sediment and oligophotic biotas can experience temperature producers during the Cenozoic, especially during the ranges of several degrees on internal-wave cycles. Paleocene-Eocene, understanding and interpreting the Those same organisms are experiencing comparable environments in which LBF were living continues to or greater changes in nutrients, organic carbon, and be both scientifically and economically important. Hallock et al. (1991) compared Paleogene evolution pCO2. For LBF and other calcifying organisms, the and extinction events in LBF and planktic latter may be particularly important as higher pCO2 in the cooler waters, when pushed into shallower depths, foraminifers, postulating that changing patterns in should result in strong saturation gradients. ocean circulation influenced nutrient regimes that were reflected in changing biotic assemblages. One Cenozoic LBF: A brief summary goal of our paper is to consider how the evolution of The end-Cretaceous extinction event (~65 Ma) was oceanic conditions during the transition from the followed by an extended carbonate depositional hiatus Greenhouse world of the early Paleogene to the (e.g., Newell 1982). Surviving shallow-water benthic Icehouse world of the Neogene influenced foraminifers with established algal symbioses (e.g., mesophotic to oligophotic habitats and biotas. miliolids) diversified, producing new lineages. New An interesting comparison can be made between symbiotic associations developed in other lineages, the history of the LBF as interpreted by Hottinger especially diatom associations with rotalid taxa (Fig. (1998) and history of ocean stratification, as indicated 4). By ~60 Ma, new benthic taxa with complex by bathymetric temperature gradients that have been morphologies were appearing (Hottinger 1998) and reconstructed for the subtropical north Pacific (Dutton by the early Eocene, large, complex nummulitids et al. 2005). Note that the diversification of the large were prominent carbonate-sediment producers in a nummulites and other Paleogene LBF occurred when variety of shelf and ramp environments (Beavington- the deep sea was warmer (Fig. 4). The apex of these Penney & Racey 2004). Extinctions eliminated the groups occurred when there was minimal temperature largest and most prominent forms of nummulitids and gradient between the thermocline and the deep sea, as 18 orthophragminids by the end of the Eocene. reflected by minimal differences between Δ O in the New taxa arose in the Oligocene, the most shells of surface-dwelling planktic foraminifers and prominently the lepidocyclinids and miogypsinids, the shells of either Subbotina, which are thermocline- though diversities, both horizontal and vertical were dwelling foraminifers, or shells of deep-sea benthic somewhat lower than during the Eocene apex foraminifers (Fig. 5). In contrast, times when bottom (Beavington-Penney & Racey 2004). Lepidocyclinids temperatures were rapidly decreasing were and other characteristic Oligo-Miocene taxa were characterized by increasing rates of extinctions of gone by the Early Pliocene. The amphisteginids, LBF (Fig. 4).

19 Hallock P (1987) Fluctuations in the trophic resource continuum: a Directions for future studies factor in global diversity cycles? Paleoceanography 2:457-471 Hallock P (1988) Interoceanic differences in foraminifera with We postulate that the history of changes in Cenozoic symbiotic algae. Proc 9th Int Coral Reef Sym 3:251-255 ocean circulation associated with changes from early Hallock P (1999) Symbiont-bearing Foraminifera. in Sen Gupta B Paleogene Greenhouse to Neogene Icehouse (ed) Modern Foraminifera, Kluwer, Amsterdam, pp 123-139 conditions are recorded in evolutionary changes in Hallock P, Glenn. EC (1986) Larger foraminifera: a tool for paleoenvironmental analysis of Cenozoic carbonate facies. LBF biotas. We recommend that future studies Palaios 1:55-64. compare details of stratification of the ocean, as Hallock P, Schlager W (1986) Nutrient excess and the demise of recorded by planktic foraminifers, with corresponding coral reefs and carbonate platforms. Palaios 1:389-398 changes in LBF assemblages. We postulate that Hallock P, Forward LB, Hansen HJ (1986) Environmental influence of test shape in Amphistegina. J Foramin Res 16:224- environmental conditions of middle and outer shelves 231 can be more thoroughly interpreted by understanding Hallock P, Premoli-Silva I, Boersma. A (1991) Similarities changes in thermocline structure of impinging oceanic between planktonic and larger foraminiferal evolutionary trends waters. Such progress will require more detailed through Paleogene paleoceanographic changes. Palaeogeogr Palaeocl 83:49-64 comparisons of planktic and LBF records, and more Hohenegger J (2005) Estimation of environmental paleogradient research on the role of internal waves on modern values based on presence/absence data: a case study using outer shelf habitats. benthic foraminifera for paleodepth estimation. Palaeogeogr Palaeocl 217:115-130 Hohenegger J (2006) The importance of symbiont-bearing benthic foraminifera for West Pacific carbonate beach environments Mar Micropaleontol 61:4-39 Hottinger L (1998) Shallow benthic foraminifera at the Paleocene- Eocene boundary. Strata ser 1, 9:61–64 Jarrett BD, Hine AC, Halley RB, Naar DF, Locker SD, Neumann AC, Twichell D, Hu C, Donahue BT, Jaap WC, Palandro D, Ciembronowicz K (2005) Strange bedfellows - a deep-water hermatypic coral reef superimposed on a drowned barrier island; southern Pulley Ridge, SW Florida platform margin. Mar Geol 214:295-307 Lear CH, Elderfield H, Wilson PA (2000) Cenozoic deep-sea temperatures and global ice volumes from Mg/Ca in benthic foraminiferal calcite. Science 287(5451):269-272 Lee JJ (1998) "Living sands" - Larger foraminifera and their endosymbiotic algae. Symbiosis 25:71-100 Lee JJ, Anderson OR (1991) Symbiosis in foraminifera. in Lee JJ, Anderson OR (eds) Biology of Foraminifera. Academic Press, London, pp. 157– 220 Leichter JJ, Genovese SJ (2006) Intermittent upwelling and Figure 5: Data from Dutton et al. (2005) showing surface-to- subsidized growth of the scleractinian coral Madracis mirabilis thermocline (Subbotina) versus surface-to-bottom Δ18O gradients, on the deep fore-reef slope of Discovery Bay, Jamaica. which indicate temperature gradients at the Shatzky Rise, Mar Ecol Progr Ser 316:95-103 subtropical North Pacific, during the Paleocene and Eocene. The Leichter JJ, Deane GB, Stokes MD (2005) Spatial and temporal bracketed time when differences between the gradients were variability of internal wave forcing on a coral reef. J Phys minimal corresponds to the apex time for Eocene nummulites and Oceanogr 35:1945-1962 other LBF according to Hottinger (1998). Mateu-Vicens G, HallockP, Brandano M (2008) Test shape variability of Amphistegina d'Orbigny 1826 as a Acknowledgement paleobathymetric proxy: application to two Miocene examples. The idea for this paper emerged from research funded by NSF in Demchuk T & Gary A (eds) Geologic Problems: Solving with BE/CBC grant CHE 0221834 and Spanish grant CGL2005-00537. Microfossils, SEPM Spec. Publ. 93:67-82. Newell ND (1982) Mass extinctions: illusions or realities? Geol. References Soc Am Spec Pap 89: 63–91. Pomar L (2001) Types of carbonate platforms, a genetic approach. Beavington-Penney SJ, Racey A (2004) Ecology of extant Basin Res 13:313-334 nummulitids and other larger benthic foraminifera: applications Pomar L, Hallock P (2008) Carbonate factories: A conundrum in in palaeoenvironmental analysis. Earth-Sci Rev 67:219-265 sedimentary geology. Earth- Sci Rev 87:134-169 Debenay J-P, Guillou J-J, Geslin E. Lesourd M (2000) Stoecker DK (1998) Conceptual models of mixotrophy in Crystallization of calcite in foraminiferal tests. planktonic protists and some ecological and evolutionary Micropaleontology 46 suppl 1:87-94. implications. Eur J Protistol 34:281-290 Dutton A, Lohmann KC, Leckie RM (2005) Insights from the Wolanski E, Colin P, Naithani J, Deleersnijder E, Golbuu Y (2004) Paleogene tropical Pacific: Foraminiferal stable isotope and Large amplitude, leaky, island-generated internal waves around elemental results from Site 1209, Shatsky Rise. Palau, Micronesia. Estuar Coast Shelf S 60:705-716 Paleoceanography 20:PA3004, doi:10.1029/2004PA001098 Hallock P (1981) Algal symbiosis: a mathematical analysis. Mar Biol 62:249-255 Hallock P (1985) Why are larger foraminifera large? Paleobiology 11:195-208

20